小波互信息及其在RR序列分析中的应用

Application of Wavelet Mutual Information in Cardiac RR Intervals Series

目的：探究信号小波成分中的有用信息,并分析心率随年龄的变化。方法：提出了从小波变换（WT）系数的延时互信息MI得出信号分析指标。以年轻（21-34岁）与年老（68-81岁）二组健康人的心电RR间期时间序列为实验数据,用db4为母小波对RR间隔进行WT,计算出MI的值,并取互信息最大值点与第一个极小值点的连线斜率的绝对值｜k｜为特征指标。结果：统计检验的结果显示：在不同尺度a=120,a=170,a=10时,小波变换系数的互信息的｜k｜都是年轻组明显大于年老组（P〈0.002）。结论：结果表明,RR间期时间序列的小波变换系数的延时互信息可反应心率变异性随年龄而减少。

Objective：To explore the useful information in wavelet components of a signal and the change of HRV with age.Methods：There exists a new method.The analysis indices of the signal are obtained from time-delayed mutual information（MI）of wavelet transform（WT） coefficients.The cardiac RR intervals series are used as experimental data collected from young（21-34 yr）and elderly（68-81 yr）groups of healthy subjects.The data analysis is performed by WT using Daubechies 4 as a mother wavelet.The characteristic indice ｜k｜ is calculated.Results：The results of statistical test show that all ｜k｜ of WT coeffi-cients at a = 170,a = 120,a = 10 are significantly higher in the young group than that in the elderly group（P〈 0.002）.Con-clusions：The time-delayed MIs of cardiac RR intervals series indicates that the heart rate variability is decreased with aging.